This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
Announcing Alteryx + Snowflake | Alteryx and Snowflake make analytics and data science fundamentally easier. With the new integrated starter kit, you can push down data prep transformations and more into Snowflake for faster data quality and analytics output. Learn More
A fashion company, founded in Los Angeles in 2002, quickly gained recognition as a designer jeans and it became a pop culture phenomenon. They entered the fashion scene before online apparel sales were common, and the brand grew increasingly popular as it was splayed across fashion magazines and sprinkled into the lyrics of hip-hop music. But with the growth of athleisure, and as online shopping continually upended the traditional brick-and-mortar retail model, their business suffered and in 2017 was just one of several retailers to seek bankruptcy protection, close stores or go out of business entirely.
Describe the business challenge or problem you needed to solve Emerging from bankruptcy and determined to reignite a relationship with its core customers, this fashion company made significant investments in data and technology, most notably investments in data science and analytics in an effort to truly understand customer shopping behavior. One problem, in particular, the company faced was reversing the trend of over-discounting items, as customers had become accustomed to deep price discounts resulting in eroding profit margin. To combat the problem, Particle planned and executed a three-phase approach.
Conduct a comprehensive data forensics audit using Alteryx Designer to assess the overall health and accuracy of the fashion company’s existing customer data.
Build a quantitatively-driven customer segmentation that could be used for ongoing marketing efforts.
Create a pricing probability model to determine which customers were most likely to pay full price for products.
Describe your working solution
Once the customer segmentation was completed, Particle went to work creating the pricing probability model. Primarily relying on the predictive tools in Alteryx Designer, Particle designed a probability model using a decision tree algorithm to identify which customers were most likely to pay full price for products. The rationale being that the company should not discount products for customers willing to pay full price. During that time, Particle created additional Alteryx workflows to help the fashion company with other needs including email address repair, fuzzy matching customer records, and data imputation methods.
Describe the benefits you have achieved
The first use of the new customer segmentation and pricing probability model was executed in a holiday marketing campaign and delivered a 275% lift in revenue compared to the same campaign the prior year. Encouraged by early success, the fashion company furthered its use of the segmentation, through A/B testing as part of a larger learning agenda Particle created.In just three months, the segmentation strategy and pricing probability model had reversed the three-year downward trend of average customer lifetime value (CLTV).